DNA testing raises ingredient declaration concerns

It behooves the petfood manufacturer to practice due diligence in ensuring that what is declared on the label is in the product, and what is not declared is not in the product.

A recent study (officially "published" in 2015, although available for purchase online now) has sparked controversy among both consumers and members of the petfood industry. It found that by polymerase chain reaction (PCR) assay (a form of DNA testing), a notable percentage of petfoods contained materials from species of animals not identified in their respective ingredient declarations. These findings are similar to those found in a previous paper from 2013 which used both PCR and microscopic methods of analysis, although that study was conducted in Europe and focused on products reportedly intended to be used for elimination trials (e.g., for food allergy diagnosis).

In addition, the more recent paper reported that some products did not contain DNA from species that were declared on the label. The authors of both papers raise questions as to the accuracy of ingredient declarations on commercial petfood labels.

Under US Food and Drug  Administration (FDA) regulations (Title 21 Code of Federal Regulations Part 501), a petfood label must bear a complete and accurate ingredient list. Unless legally exempt as an incidental additive, everything becoming a component of a food must be declared. Also, if an ingredient is not in the product, it cannot be declared. In either case, a discrepancy between the ingredient list and the actual contents constitutes a violation of law, and could subject the product to enforcement action as a misbranded, or in some cases an adulterated, food.

That said, it is not terribly surprising that the more recent study detected DNA from undeclared species in 16 of the 52 products tested (~31%). The most common discrepancy was the presence of pork DNA in products that failed to declare a possible pork-based ingredient (including "meat" or "animal" ingredients). However, one cannot conclude from this finding that any of those incidences was done intentionally or even that it constituted misbranding. Considering the typical operation of a petfood (in fact, even a human food) facility, the opportunity for cross-contact between ingredients is measurable. Further considering the sensitivity of this non-quantitative test method, bringing pork-based ingredients into a facility is likely to result in trace amounts of pork DNA being detected in other products.

With that in mind, it may be unreasonable to expect a petfood intended for the healthy individual that does not declare a particular species to be totally devoid of DNA from that species. However, what about foods intended for animals that may be sensitive to particular animal proteins? If a label specifically claims "no chicken" (rather than just failing to declare chicken in the ingredient list), how should the purchaser interpret that, and what higher standard does that manufacturer impose upon itself? It has been suggested that manufacturers of veterinary-dispensed products (e.g., therapeutic petfoods) institute greater control over processing to minimize potential cross-contact. That may be true, although the earlier study from Europe found that 10 out of 12 products tested were found to contain materials from undeclared species, despite their reportedly intended use as elimination diets (not clear if they were veterinary-dispensed or not).

Guidance by FDA with  regard to application of the Food Allergen Labeling and Consumer Protection Act (FALCPA) advises manufacturers to make efforts to mitigate the risk of cross-contact for products that do not declare the presence of major potential allergens. Some manufacturers take it a step further, and add a label disclaimer to the effect that the facility in which the product is made also makes other products that do contain one or more potential allergens. Regardless, FDA has opined that FALCPA does not apply to animal feeds or petfoods, notwithstanding that the actual wording of the amendments to the law does not make that clear.

Certainly, the consequences of exposure to a food allergen can be much more serious in humans than in dogs or cats (e.g., imminent death from anaphylactic shock versus prolongation of skin and gastrointestinal signs), so it can be argued that enforcement priority should be rightfully given to human foods. Nonetheless, failure to achieve successful long-term dietary management of a pet with a documented food allergy can be an ordeal for the animal and frustrating for its owner. Application of FALCPA would not remedy the situation, anyway, since the major allergens for dogs and cats would be largely different from what's covered by FALCPA.

More disturbing in the more recent study is the finding that in seven petfoods (13% of those tested), a declared ingredient was not found by PCR assay (e.g., no beef DNA in a product purportedly containing a beef-based ingredient). The authors documented steps taken to minimize the risk of false negatives. So, if this finding is a representative account of the prevalence of this occurrence in the marketplace, at best there appear to be some sloppy manufacturing practices going on. I understand that many ingredients may look similar upon visual inspection, so mistakes can happen, but whether deliberate or not, or whether it occurs at the manufacturer or supplier level, there's really no good excuse. Irrespective of any true safety concern or degree of enforcement priority, it's reasonable for people to expect to get what they pay for.

Some have suggested  that the reason for the findings in these studies is due to the long- and widely held belief that the regulations allow for a six-month grace period to revise labels after a formulation change. Frankly, I am not aware of any law, regulation, policy, precedent, guidance or other document written by an authoritative body that allows for this sort of transition period. Under certain circumstances, some state feed control officials might grant some time for a company to correct a label or remove an objectionable ingredient, but that does not translate to tacit permission to change formulas at whim without concurrent label changes.

Still, I would not be shocked if some in the industry have been operating under the premise that they are explicitly allowed this grace period. For the record, I advise (and have always advised) that if a manufacturer has reformulated but still has a large inventory of outdated labels, the proper means of remedy would be to apply a firmly-adhered sticker with the correct ingredient list over the old one.

Detecting discrepancies between ingredient declarations and actual contents has always been difficult, hence historically not often scrutinized by regulators unless there was probable cause. However, with more effective means of detecting these incongruities, and with the publicity this issue has received through publication of these studies, it would not be difficult to anticipate that increased monitoring and enforcement is forthcoming. The PCR methodology also provides a new tool for competitors or other third parties to publicly expose problems with petfoods.

Therefore, it behooves the petfood manufacturer to practice due diligence in ensuring that what is declared on the label is in the product, and what is not declared is not in the product. The latter aspect is especially true for those who claim the product is expressly devoid of a particular ingredient or that convey through direct or indirect means that the product is suitable for dogs or cats suffering from food allergies or intolerances.


Okuma TA, Hellberg RS. Identification of meat species in pet foods using a real-time polymerase chain reaction (PCR) assay. Food Control 50:9-17, 2015.

Ricci R, Granato A, Vascellari M, et al. Identification of undeclared sources of animal origin in canine dry foods used in dietary elimination trials. J Anim Phys Anim Nutr 97:32-38, 2013.

Page 1 of 111
Next Page